Sciweavers

103 search results - page 1 / 21
» Asymmetric Memory Fences: Optimizing Both Performance and Im...
Sort
View
CGO
2009
IEEE
13 years 11 months ago
Reducing Memory Ordering Overheads in Software Transactional Memory
—Most research into high-performance software transactional memory (STM) assumes that transactions will run on a processor with a relatively strict memory model, such as Total St...
Michael F. Spear, Maged M. Michael, Michael L. Sco...
IPPS
2003
IEEE
13 years 10 months ago
Optimizing Synchronization Operations for Remote Memory Communication Systems
Synchronization operations, such as fence and locking, are used in many parallel operations accessing shared memory. However, a process which is blocked waiting for a fence operat...
Darius Buntinas, Amina Saify, Dhabaleswar K. Panda...
CLUSTER
2009
IEEE
13 years 2 months ago
Design alternatives for implementing fence synchronization in MPI-2 one-sided communication for InfiniBand clusters
Scientific computing has seen an immense growth in recent years. The Message Passing Interface (MPI) has become the de-facto standard for parallel programming model for distribute...
Gopalakrishnan Santhanaraman, Tejus Gangadharappa,...
SC
2005
ACM
13 years 10 months ago
Making Sequential Consistency Practical in Titanium
The memory consistency model in parallel programming controls the order in which operations performed by one thread may be observed by another. Language designers have been reluct...
Amir Kamil, Jimmy Su, Katherine A. Yelick
PVM
2004
Springer
13 years 10 months ago
Minimizing Synchronization Overhead in the Implementation of MPI One-Sided Communication
The one-sided communication operations in MPI are intended to provide the convenience of directly accessing remote memory and the potential for higher performance than regular poin...
Rajeev Thakur, William D. Gropp, Brian R. Toonen